Epidemiology of Diarrhea for Chinese Dairy Calves: A Systematic Review and Meta-analysis

Calf diarrhea is one of the most serious diseases in the dairy industry, which has caused huge economic losses to the worldwide dairy industry, including China. We conducted a meta-analysis to systematical assess the prevalence and associated risk factors of diarrhea among dairy calves in China. The China Science and Technology Journal, China National Knowledge Infrastructure (CNKI), Wan Fang and Pub Med databases were systematically searched from the inception dates to April 30, 2022, using keywords of prevalence, diarrhea, calves, key factors, determinants, related factors, causes, etiology, prevalence, epidemiology and China. A meta-analysis was performed to calculate 95% Cis using random-effects models. 35 studies reporting the prevalence of dairy calves’ diarrhea in China were selected upon our inclusion criterion. The pooled diarrhea prevalence in dairy calves in China was estimated to be 27.2% (95% CI 24.2-30.2). The results of our systematic review and meta-analysis revealed that the highest prevalence rate was found in Northeast China, followed by Northwest China and East China and the lowest two regions were Northern and central China. The main cause of diarrhea in calves is bacteria, the second is virus and the third is a parasite.

dairy industry.Particularly in large-scale feeding, the incidence of diarrhea is commonly at 90%-100% and the mortality can be more than 50% at the highest status.When happening diarrhea, calves mainly show diverse clinical symptoms including depression, temperature rising, loss of appetite abdominal distention and pain, acute diarrhea, faeces with stench or mixed with bloodshot, dehydration and autotoxicosis, which caused a great influence on its survival rate, growth and development.Thus, diarrhea is generally called a newborn calf killer (Sun 2003) resulting in significant economic losses for the dairy industry.
The causes of calves' diarrhea are complex and can be roughly divided into infectious and non-infectious factors, which are closely related and complement each other.Infectious factors include viral infection, such as bovine coronavirus, bovine rotavirus and bovine viral diarrhea virus, etc., bacterial infection, including Escherichia coli, Salmonella and Shigella, etc., parasitic infection, including Cryptosporidium parvum, etc (Zhi et al., 2015).
To the best of our knowledge, the national-wide seroprevalence of diarrhea in dairy calves in China remains unknown.Therefore, we conducted this systematic review and meta-analysis to explore the epidemiology dynamics and diarrhea seroprevalence in dairy calves, which provide a theoretical basis for further extensive, comprehensive investigation and effective intervention measures according to different disease areas and different pathogens.

Search strategy
Four databases, China Science and Technology Journal, China National Knowledge Infrastructure (CNKI), Wan Fang and Pubmed, were searched for relevant literatures without date restriction.We focused on the infectious data on diarrhea of dairy calves in China and the deadline was set for April 30, 2022.The principle keywords included "prevalence", "diarrhea", "calves", "key factors", "determinants", "related factors", "causes", "etiology", "prevalence", "epidemiology" and "China".The searching process excluded literature of investigation, literature with animals other than calves (such as pigs, bison, buffaloes, beef and yaks), literature with a sample size of less than 30, literature with epidemiological results, literature with ambiguous diagnostic methods and literature outside China.

Epidemiology of Diarrhea for Chinese Dairy Calves: A Systematic Review and Meta-analysis
Population: Studies with dairy calves only.Published condition: Both published and unpublished articles were included.Study design: All observational study designs (i.e., crosssectional, case-control and cohort) reporting the prevalence of diarrhea in dairy calves were eligible for this review.

Literature screening and data extraction eligibility criteria
Data were independently extracted and recorded from each selected study by the reviewers.If there were disagreements or uncertainties on the eligibility of the study between the reviewers, it would be further assessed by other reviewers.A standardized data-collection form was used to record data as follows: the first author, publication year, pathogen, sampling mode, the number of experimental dairy calves and diarrhea dairy calves, measurement method and province.In addition, we neither contacted the authors of the original study for further information, nor identified unpublished data.The data-collection form was showed in Table 1.

Quality assessment
The quality of an eligible publication was evaluated based on criteria that Grading of Recommendations Assessment, Development and Evaluation method.The quality of publications was graded by a scoring method.In brief, when elaborating information, the score in each followed item was determined as 1 score, including target, detection method used, sampling method and subgroup classification in the study.According to this standard, one paper can be assigned into 0-4 scores.Papers with 3 and 4 scores were of high quality, with 2 scores were deemed moderate quality and those with 0 and 1 scores were validated to be low quality.Due to different detection methods in some literatures, it should be noted that there would be inconsistent scores in the same literature when investigating different pathogens.The paper with a lower score still met the criteria for inclusion in this meta-analysis, although it might not provide further analysis information.

Data processing and analysis
Data were extracted in a Microsoft Excel format and used to calculate the pooled prevalence of diarrhea in dairy calves by meta-analysis based on numerous publications.Since more heterogeneity will be expected in the qualified study, a random effect model was used to calculate and prepare the forest plot with Stata software (version 12.0, Stata Corp, College Station, Texas).The heterogeneity was predicted and the variation was assessed by using statistical methods with I2 and Cochran Q (represented as ² and p-values) statistics.Experimental statistics indicated that there existed a significant heterogeneity between studies (I2=99.0%,p<0.001).The potential sources of heterogeneity were further investigated by subgroup analysis and Metaregression analysis.The factors that caused heterogeneity in this study were examined by individual models or multivariable models.These factors included publication years (before 2010 vs after 2010 or later), pathogen types and regional differences.Point prevalence and 95% confidence interval were presented in the form of a forest map, in which the size of each box represented the weight of this study and each cross line represented a 95% confidence interval.This section might be divided into subheadings.It should provide a concise and precise description of the experimental results, their interpretation as well as the experimental conclusions that can be drawn.

Studied included
A total of 1,708 articles related to diarrhea of calves were retrieved, in which 35 articles were selected for quantitative analysis (Table 1 and Flow Chart 1), all of which are crosssectional study designs.The quality of each article was evaluated according to the article content, calves' diarrhea prevalence, research purpose and comprehensiveness of the presented data.In the 35 papers, 29 papers were scored as high quality (3 or 4 points), 5 papers as medium quality (2 points) and the remaining one as low quality (1 point) (Flow Chart 1).

Publication bias
The extent of bias in the publication in selected studies was used to assess and illustrate by the funnel plot and forest plot (Fig 1 and 2).This asymmetry indicated that there was a publication bias in selected qualified papers, which might have a significant impact on analysis results (Table 1).

Pooled prevalence of dairy calves' diarrhea in China
The results indicated that the pooled prevalence of dairy calves' diarrhea in China was 27.2% (95% confidence interval: 24.2-30.2) (Fig 2).High heterogeneity was observed in the studies (I 2 = 99.0,p<0.001).Therefore, a meta-analysis model with random effects was performed to estimate the prevalence of diarrhea in Chinese dairy calves.The metaanalysis shows that the highest prevalence was 91.7% (95% confidence interval: 85.3-98.1),while the lowest was 1.7% in other reports (0.8-2.6) Zhang et al., 2017).

Diarrhea prevalence of dairy calves caused by different pathogens in China
Bacteria, virus and parasite could all cause dairy calves' diarrhea.According to the analysis of different factors affecting diarrhea, the highest incidence of diarrhea was caused by bacteria (39.2%, 95% CI 0.30-0.48),followed by the virus (24.9%, 95% CI 0.19-0.31)and the lowest was caused by parasite (20 8%, 95% CI 0.15-0.27).The table shows publication years, pathogen causing diarrhea in calves, number of samples, positive rate, research areas and paper quality scores.The score in each followed item was determined as 1, including target, detec tion methods used, s ampling method, subgroup classification in the study.

Diarrhea prevalence of dairy calves during different periods in China
All the study date was divided into two periods: early (1992-2009) and recent (2011-2018) for subgroup analysis.During the early period, a few literature on dairy calves' diarrhea was observed, but there was a higher prevalence of 45.4% (95% CI 0.33-0.58).However, in the recent period, the diarrhea prevalence significantly decreased to 17.3% (95% CI 0.15-0.19).
In this systematic review, a meta-analysis validated that diarrhea in dairy calves was widely distributed throughout China and the average diarrhea prevalence was 27.2% from the year 2003 to 2018.Nationwide, bacteria resulted in the highest diarrhea prevalence approximate 39.2%, among which the Northeast, Northwest, East China and North China were in the forefront, with 36.2%, 32.5%, 36.5% and 30.2%, respectively.However, the prevalence of diarrhea caused by bacteria was only 6.6% in Central China, which was far lower than the national average.It might due to the central part of China was not the primary producing area of China dairy farming, thus the probability of infection was lower than other areas.Mean while, the present study showed that the main pathogen result in calves' diarrhea was Escherichia coli and its prevalence rate was 44.1%.Sun, (2016) isolated 140 strains of Escherichia coli from 420 fecal samples of diarrhea calve from six provinces and cities in northern analyzed its pathogenesis and distribution of the O serotypes.The results indicated that all the isolated strains contained 15 serotypes, where O14, O15, O101, O78 and O36 were dominant serotypes.Pathogenesis island (HIP) was found positive in 59 strains and the positive rate was 42.14% (59/140).Zhang et al., (2017), collected 72 pathological materials from different organs (intestines, lungs, urachus, real stomach, et al) of dead calves and identified 63 Escherichia coli strains.The results showed that the virulence factors were involved in 46 strains of Escherichia coli, in which 18 strains carrying CS31A gene accounted for 39.13%, indicating the CS31A gene was the dominant gene-A total of 35 strains in enterotoxigenic Escherichia coli (ETEC) were identified through typing the Escherichia coli by virulence factors, accounted for 76.09%, demonstrating that ETEC was the principle pathogenic type resulting in calves' diarrhea.Li et al., (2018) isolated 29 strains of Escherichia coli from fecal samples in 53 diarrhea calves in a cattle farm in Xinjiang province and observed that the dominant serotype was O78.Among 29 strains of Escherichia coli, 7 strains presented high virulance, accounting for 24.14%.The isolated strain was sensitive to azithromycin and the sensitive rate was 89.66%.
Meta-analysis results indicated that the diarrhea prevalence caused by the virus in Northwest and Northeast China was 26.2% and 22.2%, respectively.However, that diarrhea prevalence was remarkably decreased in East China, North China and central China, with decreased by 13%, 10% and 8.4%, respectively.The present study indicated that bovine rotavirus (BRV), bovine viral diarrhea virus (BVDV) and bovine coronavirus (BCV) were the dominant pathogens in viral diarrhea of calves and the corresponding prevalence was 32.7%, 22.2% and 3%.This showed that rotavirus was the main pathogen that caused diarrhea in calves.
In Northwest China, (Duan et al., 2009) detected rotavirus using the ELISA method in 274 fecal samples of diarrhea calves from 7 large cattle farms in Xinjiang and the average positive rate was 37.2%.Another literature reported that 172 calf diarrhea samples were collected from Xinjiang for rotavirus detection and the positive rate of the samples was as high as 72% (Lu et al., 2018).In Northeast China, Previous literature reported that the positive rate of rotavirus in diarrhea fecal samples was 57.8%, while no coronavirus was found (Wang et al., 2016).A study in Henan Province in Central China indicated that the positive rate of rotavirus and coronavirus in fecal samples of diarrhea calf was 12.24% and 4.76%, respectively (Zhao 2018). Yu, (2003) detected BVDV in serum samples of newborn calves from some cattle farms in Nanjing in East China and the positive rate was 25.96%.These studies showed that rotavirus was the most dominated pathogen in viral diarrhea calves and a general infection of rotavirus was observed in dairy-producing areas in the northwest and northeast, while the diarrhea caused by coronavirus was the lowest.From the perspective of regional differences, the risk and probability of rotavirus infection were higher in the Northwest than those in East China, North China and central China.The probable reason was that the Northwest and the Northeast were the main production areas of dairy cows in China and the number of cows was larger, which posed a threat to dairy farming in this region.From 2009 to 2018, the diarrhea rate of calves caused by rotavirus did not decrease in the main production areas of dairy cows feeding, reflecting that bovine rotavirus was not effectively controlled and there was a widespread infection by bovine rotavirus.In the long term, it is necessary to make a comprehensive intervention strategy to continuously and epidemiology monitor rotavirus infection in dairy calves.
The present results showed that the highest prevalence of diarrhea caused by the parasite was 24% in Northwest China, followed by 17.8% in Central China and 10% in North China.There were only two reports on both Northeast China and East China in the existing literature, in which the prevalence rates in Northeast China was 17.8% and 1.7%, respectively and the prevalence rates of the two cases in East China were 37% and 1.7%, respectively, quite different from each other.Therefore, more in-depth and comprehensive investigations are recommended in these Epidemiology of Diarrhea for Chinese Dairy Calves: A Systematic Review and Meta-analysis The study names included in the analysis are shown in Table 1 with their corresponding effect size and 95% confidence interval (CI).The overall effect is shown at the bottom, represented by shaded diamond.The vertical dotted line represents a risk ratio (RR) of 1, indicating no significant difference in literatures.
two areas to verify the diarrhea prevalence in dairy calves caused by parasites and to emphasize interventions to control parasite infection.Our study further found that Cryptosporidium was the main pathogen causing calves' diarrhea in parasites infection and that prevalence was 14.1%.Cryptosporidium can be transmitted by food, water and direct or indirect contact.Previous studies showed that Cryptosporidium could be mixed with rotavirus and caused diarrhea in calves (Wang et al., 2016).Dai et al., (2018) found that Cryptosporidium infection might be related to polluted water sources in dairy farms.The oocysts of Cryptosporidium are mostly discharged at night, thus they can be cleaned and disinfected in the early morning.In addition, the chlorine contained in disinfectant had a good killing effect on Cryptosporidium in drinking water and the killing rate could be greater than 99% (Liu et al., 2013).
There were also some limitations in our analysis.Firstly, some relevant research were obtained through systematic review, but these searches might not cover all studies.Secondly, the data used in this analysis considered only the publication year, geographical location or testing methods, other than potential risk factors, such as the type of cattle including free range or intensive breeding.Finally, the present study did not consider the effects of different strains and virus types on calves' diarrhea, which would be further analyzed in the future research.

CONCLUSION
In summary, diarrhea is an important disease in newborn dairy calves.The prevalence of diarrhea is different in different areas and different pathogens.In the future, more extensive surveys should be carried out and based on this to take corresponding measures to decrease the incidence and prevalence of dairy calves.

Fig 1 :
Fig 1: Funnel plot with pseudo 95% confidence interval.Funnel plot shows the incidence of diarrhea in calves, indicating that there is significant heterogeneity in the literature.

Fig 2 :
Fig 2: Forest plot of prevalence of dairy calves' diarrhea in China.